An inspection system for a vehicle underbody in an in-line process includes: a vehicle recognition unit for acquiring a vehicle ID by recognizing a vehicle entering an inspection process; a vision system that photographs the vehicle underbody through a plurality of cameras disposed under a vehicle moving direction (Y-axis) and disposed at vertical and diagonal angles along a width direction (X-axis) of the vehicle; and an inspection server that detects assembly defects of a component by performing at least one of a first vision inspection that matches an object image for each component through a rule-based algorithm or a secondary deep-learning inspection through a deep-learning engine by acquiring a vehicle underbody image photographed by operating the vision system with setting information suitable for a vehicle type and a specification according to the vehicle ID.
Legal claims defining the scope of protection. Each claim is shown in both the original legal language and a plain English translation.
3. The inspection system of claim 1, wherein the plurality of cameras are applied with an area scan camera type for correction of an inspection cycle time and an inspection position for each frame.
4. The inspection system of claim 1, wherein the plurality of cameras are applied as a global shutter type to photograph the vehicle underbody.
The inspection system is designed for capturing high-quality images of vehicle underbodies, particularly in dynamic environments such as moving vehicles on a conveyor or assembly line. The primary challenge addressed is obtaining clear, distortion-free images of the underbody, which often contains critical components like exhaust systems, suspension parts, and structural elements. Traditional imaging methods, especially with rolling shutters, can introduce motion blur and distortion when capturing fast-moving or unevenly lit surfaces. The system employs multiple cameras configured with global shutter technology to photograph the vehicle underbody. Global shutters capture the entire frame simultaneously, eliminating the distortion and artifacts caused by rolling shutters in high-speed or uneven lighting conditions. The cameras are strategically positioned to ensure comprehensive coverage of the underbody, capturing detailed images of all relevant components without motion blur. This setup allows for precise inspection, enabling automated defect detection, quality control, and maintenance tracking in automotive manufacturing or inspection processes. The system may also include additional features such as lighting control, image processing algorithms, and data integration with inspection databases to enhance accuracy and efficiency.
5. The inspection system of claim 1, wherein the vision system includes an LED plate light configured on an upper surface of the base and an LED ring light configured on an installation surface of each of the tilting cameras, and each light filters a diffuse reflection through a polarizing filter.
6. The inspection system of claim 1, wherein the vision system adjusts a tilting angle (.theta.) of each of the tilting cameras through a tilting camera mounting unit that includes at least one servo motor and changes setting positions in up/down and left/right directions.
7. The inspection system of claim 6, wherein the vision system further includes a vertical elevator for vertically changing a position of the base on which the plurality of cameras are disposed.
The inspection system is designed for high-precision visual inspection of objects, particularly in industrial or manufacturing settings where detailed imaging is required. The system addresses the challenge of capturing comprehensive visual data from multiple angles and heights without physical movement of the object being inspected. The core of the system includes a base supporting a plurality of cameras arranged to capture images of an object from different perspectives. These cameras are positioned to provide overlapping or complementary fields of view, ensuring thorough coverage of the object's surface. The vision system further incorporates a vertical elevator mechanism that adjusts the vertical position of the camera base. This allows the cameras to inspect objects of varying heights or to focus on specific sections of a tall object by moving the entire camera array up or down. The system may also include additional features such as lighting control, image processing algorithms, or mechanical positioning systems to enhance inspection accuracy and efficiency. The vertical elevator ensures flexibility in adapting to different inspection scenarios without requiring manual adjustments, improving automation and reducing setup time. The overall design aims to streamline the inspection process while maintaining high-resolution imaging capabilities.
8. The inspection system of claim 1, wherein the vision system is installed so as to move back and forth according to an equipment environment through a front and rear moving device of a linear motion (LM) guide type installed at the lower part.
The inspection system is designed for automated visual inspection in industrial or manufacturing environments. The system addresses the challenge of accurately inspecting objects or components while accommodating varying equipment layouts and spatial constraints. A key feature is a vision system that can dynamically adjust its position to optimize inspection accuracy and coverage. This is achieved through a front and rear moving device of a linear motion (LM) guide type installed at the lower part of the system. The LM guide allows precise, controlled movement back and forth, enabling the vision system to adapt to different equipment environments. The system may also include a base structure supporting the vision system and the moving device, ensuring stability during operation. The vision system itself may incorporate imaging sensors, lighting, and processing components to capture and analyze visual data. The adjustable positioning capability enhances flexibility, allowing the system to inspect objects from optimal angles and distances, improving inspection reliability and efficiency. The LM guide mechanism ensures smooth, repeatable motion, which is critical for consistent inspection results in automated production lines. This design is particularly useful in environments where fixed inspection systems would be impractical due to space limitations or varying inspection requirements.
10. The inspection system of claim 1, wherein the controller converts the object image to a grayscale, compares an area feature value for each angle through template matching, extracts a matching score according to the comparison, and determines that it is defective if the matching score is less than a predetermined reference value.
This invention relates to an inspection system for detecting defects in objects using image analysis. The system addresses the challenge of accurately identifying defects in objects by analyzing their visual characteristics through a structured process. The inspection system includes an imaging device that captures an image of an object and a controller that processes the image to detect defects. The controller converts the captured object image to grayscale to standardize the visual data. It then compares an area feature value for each angle of the object through template matching, a technique that aligns the object image with a reference template to identify deviations. The system extracts a matching score based on this comparison, quantifying how closely the object matches the reference. If the matching score falls below a predetermined reference value, the controller determines that the object is defective. This approach ensures consistent and automated defect detection by leveraging image processing and pattern recognition. The system is particularly useful in manufacturing and quality control applications where precise defect identification is critical.
11. The inspection system of claim 10, wherein the controller divides the region of interest into a plurality of regions defined according to a characteristic of the component, distinguishes the divided regions with a label, and compares a label ratio of each region divided by the label with a reference ratio to determine whether the component is defective in the assembly, and the controller compares label ratios in the region divided by the label with each other to determine whether the component is defective in the assembly from a ratio that changes when any one label ratio is omitted.
14. The inspection system of claim 12, wherein the plurality of cameras are applied with an area scan camera type for correction of an inspection cycle time and an inspection position for each frame.
15. The inspection system of claim 12, wherein the plurality of cameras are applied as a global shutter type to photograph the vehicle underbody.
The inspection system is designed for capturing high-quality images of vehicle underbodies, particularly in dynamic environments such as moving vehicles on a production line. The system addresses challenges in obtaining clear, distortion-free images of vehicle underbodies, which often suffer from motion blur, lighting inconsistencies, and uneven surfaces. The system includes a plurality of cameras positioned to capture images of the underbody from multiple angles, ensuring comprehensive coverage. These cameras are configured as global shutter types, which synchronize their exposure to capture the entire frame simultaneously, eliminating rolling shutter artifacts and motion blur that can occur with traditional rolling shutter cameras. The system may also incorporate lighting mechanisms to enhance visibility and reduce shadows, as well as image processing algorithms to stitch and analyze the captured images for defects or quality control purposes. The global shutter design ensures that all pixels in the image are exposed at the same time, providing sharp, undistorted images even when the vehicle is in motion. This configuration is particularly useful in automotive manufacturing and inspection applications where precision and speed are critical.
16. The inspection system of claim 12, wherein the vision system includes an LED plate light configured on an upper surface of the base and an LED ring light configured on an installation surface of each of the tilting cameras, and each light filters a diffuse reflection through a polarizing filter.
17. The inspection system of claim 12, wherein the vision system adjusts a tilting angle (.theta.) of each of the tilting cameras through a tilting camera mounting unit that includes at least one servo motor and changes setting positions in up/down and left/right directions.
18. The inspection system of claim 17, wherein the vision system further includes a vertical elevator for vertically changing a position of the base on which the plurality of cameras are disposed.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
December 1, 2020
November 22, 2022
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.